This manuscript is a preprint. This manuscript has been submitted to Journal of the Geological

Society, having previously been peer-reviewed. Subsequent versions of this manuscript may have

different content. If accepted, the final version of this manuscript will be available via the ‘Peer-

reviewed Publication DOI’ link via this webpage. Please feel free to contact any of the authors

directly or to comment on the manuscript. We welcome feedback!

1

Are magnetic stripes on the Cuvier Abyssal Plain (offshore NW Australia) 1

diagnostic of oceanic crust? 2

3

Running title: Origin of the Cuvier Abyssal Plain 4

5

Matthew T. Reeve1, Craig Magee1,2*, Ian D. Bastow1, Carl McDermott1, Christopher A.-L. 6

Jackson1, Rebecca E. Bell1, Julie Prytulak3 7

8

1Basins Research Group (BRG), Department of Earth Science and Engineering, Royal School 9

of Mines, Prince Consort Road, Imperial College London, SW7 2BP, England, UK. 10

2School of Earth and Environment, University of Leeds, Leeds, LS2 9JT, UK. 11

3Department of Earth Sciences, University of Durham, DH1 3LE, UK 12

13

*Correspondence (c.magee@leeds.ac.uk) 14

15

Abstract 16

Magnetic stripes have long been used to define the presence and age of oceanic crust. 17

However, continental crust heavily intruded by magma can record magnetic reversals akin to 18

those observed in oceanic crust. We re-evaluate the nature of the Cuvier Abyssal Plain 19

(CAP), offshore NW Australia, which hosts magnetic stripes and has previously been defined 20

as oceanic crust. We use magnetic, 2D seismic reflection, and geochemical data to test 21

whether the CAP structure and composition is consistent with unambiguous oceanic crust. 22

We show chemical data from a basalt within the CAP, previously described as displaying an 23

enriched MORB-like signature, actually contains evidence of contamination by continental 24

material. We also recognise seaward-dipping reflector (SDR) sequences across the CAP. 25

2

Borehole data from overlying sedimentary rocks suggests these SDRs were emplaced in a 26

shallow-water (<200 m depths) or sub-aerial environment. Our results indicate the CAP may 27

not be unambiguous oceanic crust. Instead, we suggest the CAP could comprise a spectrum 28

of heavily intruded continental crust (akin to present-day Ethiopia) through to fully oceanic 29

crust, recording the evolution from continental rifting to progressively magma-dominated, 30

sub-aerial to shallow-water extension. Our work supports suggestions that magnetic reversals 31

may not be truly diagnostic of oceanic crust. 32

33

Supplementary material: Enlarged and uninterpreted versions of the magnetic data and 34

seismic reflection lines are available at. 35

36

Development of magnetic reversal anomalies (stripes) during oceanic crust formation is 37

fundamental to modern plate tectonic theory (e.g., Vine & Matthews 1963). Where magnetic 38

stripes occur adjacent to passive continental margins, they are commonly interpreted to mark 39

a basin’s oldest, unambiguous oceanic crust. Such stripes have also historically been used to 40

define the continent-ocean boundary (e.g., Talwani & Eldholm 1973; Rabinowitz & 41

LaBrecque 1979; Veevers 1986; Eagles et al. 2015): here we define unambiguous oceanic 42

crust as comprising layers of pillow basalts, sheeted dikes, and gabbro formed during deep-43

marine (≳2 km water depths) seafloor spreading (e.g., McDermott et al. 2018). However, 44

continental break-up can produce broad, complex zones that have structural and geochemical 45

traits lying somewhere between unambiguous continental crust and unambiguous oceanic 46

crust (e.g., Skogseid et al. 1992; Symonds et al. 1998; Planke et al. 2000; Skogseid et al. 47

2000; Direen et al. 2007; Bastow & Keir 2011). The crustal affinity of such break-up zones is 48

accordingly difficult to distinguish (e.g., Eagles et al. 2015). Where these so-called continent-49

ocean transition zones (COTZs) occur, as opposed to an abrupt continent-ocean boundary, 50

3

magnetic stripes within the adjacent unambiguous oceanic crust are expected to mark the 51

outer limit of the COTZ (e.g., Pickup et al. 1996; Direen et al. 2007; Eagles et al. 2015; Paton 52

et al. 2017; Peron-Pinvidic et al. 2019). Using magnetic stripes to recognise and accurately 53

map the extent and age of oceanic crust has proved critical to palinspastic and plate kinematic 54

reconstructions (e.g., Heine & Müller 2005; Eagles et al. 2015; Causer et al. 2020). Yet linear 55

magnetic stripes akin to those hosted by unambiguous oceanic crust have recently been 56

identified along: (i) the onshore Afar Rift, Ethiopia in heavily intruded continental crust 57

(Bridges et al. 2012), an area expected to eventually become a COTZ assuming full seafloor 58

spreading develops; (ii) magma-poor passive margin COTZs offshore Iberia and 59

Newfoundland, where magnetic anomalies are recorded by magmatic intrusion into exhumed 60

and serpentinised mantle prior to break-up (Bronner et al. 2011); (iii) part of the magma-rich 61

passive margin COTZ offshore NW Australia (i.e. the Gascoyne margin; Direen et al. 2008); 62

and (iv) offshore South America where the margin comprises ‘magmatic crust’ wholly 63

comprised of new igneous material, which differs from normal oceanic crust in that it formed 64

via sub-aerial and/or shallow-water extension, not deep-marine spreading (Collier et al. 2017; 65

McDermott et al. 2018). These observations suggest magnetic stripes could develop within 66

non-oceanic crust, which questions whether they can be confidently used to diagnose seafloor 67

spreading, detect continent-ocean boundaries, and be used as hard constraints on palinspastic 68

and plate kinematic reconstructions (Rooney et al. 2014). 69

Given recent studies have shown magnetic stripes may not be diagnostic of 70

unambiguous oceanic crust formed during deep-marine seafloor spreading (e.g., Direen et al. 71

2008; Bridges et al. 2012; Collier et al. 2017; McDermott et al. 2018), it is worth re-72

evaluating the nature of previously defined oceanic crust adjacent to passive margins. Here, 73

we test the origin of the Cuvier Abyssal Plain (CAP), offshore NW Australia through an 74

integrated analysis of 2D seismic reflection data from the CAP, coupled with a re-75

4

examination of published chemical data (Fig. 1). The CAP hosts well-developed magnetic 76

anomalies distributed about an inferred spreading centre along the Sonne Ridge (Fig. 1B) 77

(e.g., Robb et al. 2005). These anomalies, coupled with observations from seismic refraction 78

data, have previously been used to suggest that the CAP comprises unambiguous oceanic 79

crust (Fig. 1) (e.g., Falvey & Veevers 1974; Larson et al. 1979; Robb et al. 2005; MacLeod et 80

al. 2017). Basalts dredged from the Sonne Ridge have been interpreted to reflect an enriched 81

MORB-like chemistry, supporting the inference that the CAP crust is oceanic (Crawford & 82

von Rad 1994; Dadd et al. 2015). The CAP has thus previously been defined as oceanic crust 83

in all regional and global palinspastic and palaeogeographic reconstructions involving NW 84

Australia (e.g., Heine & Müller, 2005; Gibbons et al., 2012). 85

Within our seismic reflection data we recognise multiple packages of up to ~3 km 86

thick seaward-dipping reflector (SDR) sequences, which cumulatively extend at least ~300 87

km oceanwards from the continental shelf. These SDR packages are likely dominated by 88

lavas and occasionally span several well-defined, broadly linear magnetic anomalies. 89

Lithological and biostratigraphic data from the Deep Sea Drilling Project (DSDP) Site 263 90

borehole indicate sedimentary strata above these SDRs were deposited in neritic 91

environments (<200 m water depths), implying lava emplacement was sub-aerial or shallow-92

water. Our reinterpretation of chemical data also indicates basalts from the Sonne Ridge 93

contain a continental signature (e.g., Robb et al. 2005; MacLeod et al. 2017). We suggest that 94

the CAP may not be unambiguous oceanic crust and could instead comprise a spectrum of 95

crustal types, ranging from heavily intruded continental crust to fully magmatic crust 96

generated during sub-aerial or shallow-water extension. We consider the CAP may define a 97

COTZ with an outer-limit >500 km oceanwards from the previously defined passive margin 98

boundary. Our interpretation has important implications for palinspastic and plate kinematic 99

reconstructions involving the NW Australian margin, as well as for heat flow and basin 100

5

modelling studies. More generally, or study supports the arising hypothesis that magnetic 101

stripes may not be a unique feature of oceanic crust. 102

103

Continent-Ocean Transition Zones 104

Where we can study active, magma-rich continental break-up onshore in Ethiopia, the 105

processes driving development of a possible future COTZ during the latter stage of rifting 106

(i.e. prior to rupture) involve the localisation of extension into narrow (~20 km wide) zones 107

that progressively become dominated by dyke intrusion (e.g., Ebinger & Casey 2001; 108

Keranen et al. 2004; Mackenzie et al. 2005; Maguire et al. 2006; Daniels et al. 2014). In these 109

sub-aerial and so-called ‘magmatic segments’, ≳50% of the crust may comprise new, 110

intruded mafic material (Daniels et al., 2014). Along volcanic or magma-rich passive 111

margins, COTZs commonly contain seismically isotropic, fast (>7 km s-1) crust, overlain by 112

SDR sequences of mafic volcanic products (e.g. lava flows) interbedded with sedimentary 113

rocks emplaced within sub-aerial or shallow-water environments (e.g., Eldholm et al. 1989; 114

Larsen & Saunders 1998; Symonds et al. 1998; Menzies et al. 2002). The COTZ crust 115

underlying these SDR-bearing domains comprises a spectrum of stretched and heavily 116

intruded continental crust (e.g., Eldholm et al. 1989) to ‘magmatic crust’ (also termed 117

‘igneous’ crust), which record processes similar to those active within Ethiopian magmatic 118

segments (e.g., Collier et al. 2017; Paton et al. 2017; McDermott et al. 2018). Magmatic crust 119

is structurally similar to oceanic crust but instead formed during sub-aerial or shallow-water 120

extension along a magmatic segment, rather than deep-marine spreading at a mid-ocean ridge 121

(e.g., Collier et al. 2017; Paton et al. 2017; McDermott et al. 2018); i.e. although the 122

formation of magmatic crust requires the prior break-up of continental crust, we note that this 123

event may not coincide with full continental lithospheric rupture, which necessarily involves 124

the mantle lithosphere. 125

6

Observations from rifted margins and active rifts suggest that COTZs at magma-rich 126

passive margins are marked by a compositional and structural spectrum, bounded by 127

unambiguous continental and oceanic crust end-members (Fig. 2). From the landward limit of 128

COTZs, we expect the proportion of magma intruded into continental crust to increase 129

oceanwards (Fig. 2). As dyking localises, eventually no continental crust will remain (i.e. 130

break-up of continental crust), and the COTZ will solely comprise igneous intrusions and 131

extrusions emplaced along magmatic segments during sub-aerial or shallow-water extension 132

(i.e. magmatic crust; Fig. 2) (e.g., Paton et al. 2017; McDermott et al. 2018). Because of 133

uncertainties in data resolution and interpretation mean, it is often difficult to constrain the 134

progression from intrusion of continental crust to the onset of magmatic crust emplacement. 135

We therefore combine these domains and refer to them both simply as a COTZ (Fig. 2). 136

Diminishment of the buoyant support maintaining these dense, sub-aerial or shallow-water 137

magmatic segments will promote their subsidence (e.g., Corti et al. 2015; McDermott et al. 138

2018). As these magmatic segments subside to water depths of ≳2 km, plate-spreading drives 139

the generation of unambiguous oceanic crust, comprising layers of pillow basalts, sheeted 140

dykes, gabbro, and oceanic mantle lithosphere (e.g., McDermott et al. 2018); i.e. full 141

continental lithospheric rupture has occurred by this point. Across COTZs and into 142

unambiguous oceanic crust, we may therefore expect an oceanwards reduction in the 143

continental signature of magma chemistry as they become more MORB-like (Fig. 2). 144

145

Geological Setting 146

Crustal Structure and Age 147

The North Carnarvon Basin and South Carnarvon Basin form part of the NW Australian 148

magma-rich passive margin, bound by the Argo Abyssal Plain to the north, and the Gascoyne 149

Abyssal Plain and Cuvier Abyssal Plain (CAP) to the west (Fig. 1A) (Longley et al. 2002; 150

7

Stagg et al. 2004). Basin formation occurred during multiple phases of Permian-to-Late 151

Jurassic rifting, culminating in Early Cretaceous break-up of the Gascoyne and Cuvier 152

margin rift segments (Fig. 3A) (Longley et al. 2002). We sub-divide the study area into the 153

~400 km wide Gascoyne and 180 km-wide Cuvier margin sectors, separated by the NW-154

trending Cape Range Fracture Zone (Fig. 1A). The CAP is bound to the SW by the Wallaby 155

Plateau and Wallaby Saddle (Fig. 1A). 156

157

Previously interpreted continent-ocean transition zones (COTZs) 158

A 200–250 km wide COTZ lies northwest of the continental Exmouth Plateau (i.e. the Gallah 159

Province; Fig. 1A). This COTZ comprises 2–5.5 km thick SDRs and high-velocity lower 160

crust (Direen et al. 2008). The Gallah Province COTZ preserves magnetic stripes formed 161

during chrons M10N–M5n (~136–131 Ma; Valanginian-to-Hauterivian), with interpreted 162

oceanic crust (Gascoyne Abyssal Plain) emplaced onwards from chron M3r (~130 Ma, 163

Hauterivian; Figs 1B and 3A) (Direen et al. 2008). Along the Cuvier Margin, beneath the 164

modern continental slope, seismically imaged SDR sequences are interpreted to mark a 50–165

70 km wide COTZ (e.g., Figs 1A and C) (Hopper et al. 1992; Symonds et al. 1998). 166

167

Cuvier Abyssal Plain 168

Adjacent to the Cuvier Margin, the CAP lies ~5 km below sea level and comprises a ~1–3.3 169

km thick sedimentary sequence overlying ~6–10.5 km thick crystalline crust (e.g., Fig. 1C) 170

(Hopper et al., 1992). Based on recognition of magnetic chrons M10N–M5 within the CAP, it 171

has been interpreted as oceanic crust emplaced initially at ~136 Ma (i.e. Valanginian; Figs 1B 172

and 3A) (Falvey & Veevers 1974; Larson et al. 1979). The presence of chrons M10N–M5 173

within the Gallah Province COTZ (Direen et al. 2008) and the inference that seafloor 174

spreading began in the CAP at ~136 Ma, suggest that continental rupture and seafloor 175

8

spreading adjacent to the Cuvier Margin occurred ~5–6 Myr before the Gascoyne Margin 176

(Fig. 3A) (Falvey & Veevers 1974; Larson et al. 1979). Critically, the occurrence of magnetic 177

stripes in the non-oceanic (i.e. COTZ) crust of the Gallah Province questions whether the 178

presence magnetic stripes in the CAP should be used to classify it as oceanic crust (e.g., 179

Direen et al. 2008; Bridges et al. 2012; Collier et al. 2017; McDermott et al. 2018). 180

181

Origin of the Sonne and Sonja ridges 182

Under the assumption that the CAP comprises oceanic crust, the ~175 km long Sonne Ridge 183

and ~100 km long Sonja Ridge, which extend into the Wallaby Plateau (Fig. 1A), have been 184

interpreted as probable extinct oceanic spreading ridges (e.g., Mihut & Müller 1998; Robb et 185

al. 2005; MacLeod et al. 2017); in this model, spreading is interpreted to have jumped from 186

the Sonne Ridge to the Sonja Ridge at ~131.7 Ma (Hauterivian). Geochemical analyses of a 187

basalt dredged from the Sonne Ridge along its extension into the Wallaby Plateau, suggest it 188

has a slightly enriched MORB-like signature, supporting the inference that the Sonne Ridge 189

is an oceanic spreading centre (Dadd et al. 2015). An alternative interpretation forwarded for 190

the Sonne Ridge is that it represents a ‘pseudofault’ (i.e. an apparent offset in magnetic 191

stripes formed by ridge jumps; Hey 1977); this interpretation is based on changes in gravity 192

intensity across the structure and the possible termination of the Cape Range Fracture Zone 193

directly north of the ridge (Gibbons et al. 2012). In their model, Gibbons et al. (2012) define 194

a different oceanic spreading centre ~100 km to the SE and parallel to the Sonne Ridge (red 195

dashed line in Figs 1A and B). However, we note that Gibbons et al. (2012) use the COB of 196

Heine & Müller (2005) to define the termination of the Cape Range Fracture Zone. This COB 197

interpretation does not consider the Gallah Province is a COTZ and, hence, does not include 198

the required north-westward extension of the fracture zone beyond the seaward limit of the 199

Exmouth Plateau (Heine & Müller 2005; Robb et al. 2005; Direen et al. 2008). Overall, given 200

9

the repetition of magnetic stripe sequences either side of the Sonne Ridge, and to a lesser 201

extent the Sonja Ridge, we favour their interpretation as spreading ridges and thus use the 202

interpreted magnetic chron configuration of Robb et al. (2005). This magnetic chron 203

configuration is centred on the Sonne Ridge and predicts half-spreading rates adjacent to both 204

Gascoyne and Cuvier margins were similar during chrons M10–M5 (~4.5 cm/yr), decreasing 205

to ~3 cm/yr by chron M3 (Robb et al. 2005). 206

207

The Wallaby Plateau and Wallaby Saddle 208

The Wallaby Plateau is a large bathymetric high (Fig. 1A), containing up to ~7.5 km thick 209

sequences of volcanic and sedimentary rocks, which are typically expressed in seismic 210

reflection data as packages of diverging and dipping reflections that appear similar to SDRs 211

(e.g., Colwell et al. 1994; Daniell et al. 2009; Stilwell et al. 2012; Olierook et al. 2015). 212

Interpretation of seismic reflection and magnetic data, coupled with chemical, 213

geochronological, and biostratigraphic analyses of dredge samples, suggests the Wallaby 214

Plateau probably comprises ~124 Myr old, continental flood basalts and interbedded 215

sedimentary strata emplaced on a fragment of extended continental crust (see Olierook et al. 216

2015 and references therein). Between the Wallaby Plateau and the Australian continent is 217

the Wallaby Saddle, a bathymetric low containing SDRs but no magnetic stripes, interpreted 218

by Symonds et al. (1998) to comprise ‘transitional’ crust (Figs 1A and B). The Wallaby 219

Plateau and Wallaby Saddle seemingly preserve a range of crustal types typical of a COTZ, 220

but not unambiguous continental crust or unambiguous oceanic crust. Similarities between 221

their structure and that of the CAP may therefore imply the latter is not truly oceanic crust. 222

223

Sedimentary Cover on the Cuvier Abyssal Plain 224

10

The top of the crystalline basement within the CAP corresponds to a high-amplitude 225

reflection in seismic data, which is overlain by a ~1–3.3 km thick, sedimentary succession 226

broadly comprising sub-horizontal reflections (e.g., Fig. 1C) (e.g., Veevers & Johnstone 227

1974; Hopper et al. 1992). Biostratigraphic and lithological data for the sedimentary cover 228

are available from the DSDP Site 263 borehole, which was drilled in 1972 and terminates 229

~100–200 m above the basement (Figs 1A and 3B) (e.g., Bolli 1974; Scheibnerová 1974; 230

Wiseman & Williams 1974; Holbourn & Kaminski 1995). From a depth of 200 m below 231

seabed to the base of the borehole (746 m), the sedimentary strata intersected at DSDP Site 232

263 comprise black claystones; between depths of ~475–746 m these claystones are silty and 233

contain abundant kaolinite (Fig. 3B) (Robinson et al. 1974; Compton et al. 1992). Within the 234

lowermost ~20 m of the drilled sequence, and in thin units at around ~500 m depth, the 235

siltstones occur that are poorly sorted, and contain angular quartz grains (Fig. 3B) (Robinson 236

et al. 1974). Analyses of benthic foraminifera from DSDP Site 263 suggest the black, 237

kaolinitic claystones spanning ~475–746 m are likely Hauterivian-to-Middle Barremian, 238

passing upwards into Albian-to-Aptian black claystones (Fig. 3B) (Holbourn & Kaminski 239

1995); these age ranges are supported by dinoflagellate distributions and carbon isotope 240

stratigraphy (Wiseman & Williams 1974; Oosting et al. 2006). A gradual upwards transition 241

from coarsely to finely agglutinated foraminifera species, coupled with an upwards increase 242

in the scarcity of shallow-water taxa (e.g., Hyperamina spp.) and a corresponding decrease in 243

grain size, suggests that the Hauterivian-to-Middle Barremian strata record deepening neritic 244

(i.e. <200 m water depth) conditions (e.g., Fig. 3B) (Robinson et al. 1974; Veevers & 245

Johnstone 1974; Holbourn & Kaminski 1995; Oosting et al. 2006). Sedimentary rocks 246

recovered from the Pendock-1 borehole, which is located on the current continental shelf, are 247

sedimentologically similar and of comparable age to those penetrated in DSDP 263 (Veevers 248

& Johnstone 1974; Holbourn & Kaminski 1995). These similarities to Pendock-1 suggest that 249

11

the Hauterivian-to-Middle Barremian strata sampled by DSDP Site 263 can broadly be 250

correlated to the Winning Group of the North and South Carnarvon basins (Figs 1A and 3) 251

(Veevers & Johnstone 1974; Holbourn & Kaminski 1995). 252

253

Dataset and methodology 254

Seismic reflection data 255

To assess the crustal structure of the CAP and surrounding areas, we interpret seven 2D 256

seismic lines from four, pre-stack time-migrated reflection surveys (Fig. 1A) (see 257

Supplementary Table 1 for acquisition and processing details for each survey). Seismic lines 258

EW0113-5, EW0113-6, and repro-n303 are each >400 km long and extend from the 259

continental shelf up to ~300 km into the CAP from the currently defined COTZ (i.e. they are 260

orthogonal to the structural trend of the margin); EW0113-5 and EW0113-6 span the mapped 261

area of SDRs in the Cuvier COTZ and the location of the extinct spreading centre predicted 262

by Gibbons et al., (2012), whereas repro-n303 images the Sonne Ridge (Fig. 1A). Due to 263

extreme amplitude contrasts between the shallow and deep sections of the original migrated, 264

EW0113 data, we applied a time-dependent gain filter and root filter to improve amplitude 265

balance and enhance deep reflectivity (see supplementary information for details). Lines 266

s135-05, s135-08, and s310-59 image the inferred ‘transitional’ crust of the Wallaby Saddle 267

and the intruded continental crust of the Wallaby Plateau (e.g., Symonds et al., 1998; 268

Goncharov and Nelson, 2012; Olierook et al., 2015). The NE-trending seismic line s135-11 269

was also interpreted as ties together the margin-orthogonal seismic lines and provides a 270

margin-parallel image of the southernmost Exmouth Plateau continental crust, the CAP, and 271

the Wallaby Plateau (Fig. 1A). 272

Although time-migrated seismic reflection data allows us to qualitatively and 273

quantitatively characterise crustal structure, seismic velocity information is required to 274

12

convert depth information from seconds two-way time (TWT) to metres. To provide context 275

for the thicknesses and depths of some discussed structures, we depth-converted the 276

EW0113-5 and EW0113-6 seismic data using interval velocities derived from ocean-bottom 277

seismometer (OBS) data (Table 1) (Tischer 2006). The OBS array was co-located with 278

seismic line EW0113-6, which is located ~70 km along-strike from line EW0113-5 (Fig. 1A); 279

the geological structure imaged in line EW0113-6 is very similar to that of EW0113-5, 280

supporting the use of velocities from EW0113-6 to depth convert both lines. As velocity data 281

from across the Wallaby Plateau and Wallaby Saddle is limited (Goncharov & Nelson 2012), 282

and because along-strike variation in geology will likely promote changes in the velocity 283

structure, lines s135-05, s135-08, and s310-59 are presented in time. For easier comparison 284

between seismic data from the CAP and the Wallaby Plateau and Wallaby Saddle, we do not 285

depth-convert repro-n303 or s135-11. Interpretation of reflection configurations (e.g., dip 286

values) in time-migrated data are only qualitative, and may change if depth-converted. 287

288

Magnetic data 289

To examine the regional magnetic anomalies, we utilise the EMAG2v2 and EMAG2v3 Earth 290

Magnetic Anomaly Grids (Maus et al. 2009; Meyer et al. 2017). EMAG2v2 is a 2 arc min 291

resolution grid derived from marine, airborne, and satellite magnetic data, but uses a priori 292

information to interpolate magnetic anomalies in areas where data gaps are present (Fig. 4A) 293

(Maus et al. 2009; Meyer et al. 2017). In contrast, EMAG2v3 uses more data points to derive 294

magnetic anomaly maps but assumes no a priori information (Fig. 4B) (Meyer et al. 2017). 295

In ocean basins with a relatively poor coverage of magnetic data available, such as the CAP, 296

clear linear magnetic anomalies in EMAG2v2 thus typically appear poorly developed or are 297

absent in EMAG2v3 (cf. Figs 4A and B) (Meyer et al. 2017). This difference in the presence 298

and appearance of linear magnetic anomalies between grids is because (assumed) knowledge 299

13

of seafloor spreading processes was incorporated into, and therefore influenced, interpolation 300

during construction of the EMAG2v2 grid (Maus et al. 2009; Meyer et al. 2017). Importantly, 301

the apparent reduction in magnetic stripes observed in EMAG2v3, compared to EMAG2v2 302

(Figs 4A and B), does not necessarily mean these features are absent, but rather that the 303

available data is insufficient to unambiguously confirm their presence in non-directionally 304

gridded data such as EMAG2v3 (Meyer et al. 2017). Comparing the EMAG2v2 and 305

EMAG2v3 grids, coupled with shiptrack magnetic data (Robb et al. 2005), allows us to 306

interrogate the magnetic architecture of the CAP (cf. Meyer et al. 2017). In particular, we use 307

EMAG2v2 to interpret possible magnetic chrons, picked on positive peaks (Fig. 1B), and 308

attempt to correlate them to the EMAG2v3 grid and shiptrack magnetic data. From these 309

comparisons, we tied interpreted magnetic stripes to seismic line EW0113-5, EW0113-6, and 310

repro-n303 using the synthetic profiles of Robb et al. (2005). To update the absolute ages of 311

the interpreted magnetic anomalies (Robb et al. 2005), we use the time-calibrated, magnetic 312

polarity reversal sequence of Gradstein & Ogg (2012). 313

314

Geochemical data 315

To evaluate whether the Sonne Ridge is an extinct seafloor spreading centre (e.g., Mihut & 316

Müller 1998; Robb et al. 2005) consisting of oceanic crust with a MORB or MORB-like 317

affinity along its length, we examine chemical data from a dredged basalt lava sample 318

collected along its extension into the Wallaby Plateau (i.e. Site 57 - sample 057DR051A; 319

diamond 57 in Fig. 1A) (Daniell et al. 2009; Dadd et al. 2015; Olierook et al. 2015). We 320

compare the Sonne Ridge sample to two samples collected from near the south-western 321

margin of the Wallaby Plateau (diamonds 55 and 52 in Fig. 1A) (i.e. Site 55 - samples 322

055BS004A and 055BS004B) (Dadd et al. 2015). Two Wallaby Plateau basalts dated from 323

Site 52 (Fig 1A) yield plagioclase 40Ar/39Ar plateau ages of 125.12±0.9 Ma and 123.80±1.0 324

14

Ma, whereas two analyses of the Sonne Ridge sample yielded less precise ages of 120±14 Ma 325

and 123±11 Ma (Olierook et al. 2015). 326

327

Results 328

Reflection seismology 329

Cuvier Abyssal Plain 330

We interpret a prominent, continuous, high-amplitude seismic reflection across the CAP; this 331

represents the interface between crystalline rock and overlying sedimentary strata (e.g., Figs 332

1C and 5). The Moho was picked at the base of a sub-horizontal zone of moderate-to-high-333

amplitude, discontinuous seismic reflections, and is broadly flat-lying at ~16–17 km or ~10 s 334

TWT (Fig. 5). On EW0113-5, the Moho appears to become shallower oceanwards (to ≤14 335

km), although our interpretation of repro-n303 suggests it may deepen again beneath the 336

Sonne Ridge (Figs 5A and D). Overall, the crystalline crust is ~8–10 km (~3–5 s TWT) thick 337

(Figs 1C and 5). On EW0113-5 and EW0113-6, there is no clear evidence for the spreading 338

ridge interpreted by Gibbons et al., (2012), which is expected to occur within magnetic chron 339

M9n (Figs 1A and 5). 340

Across the CAP, the ~1–3 km-thick, uppermost crystalline crustal layer comprises a 341

layered, moderate- to high-amplitude seismic facies (SF1; Fig. 5). On NW-trending seismic 342

lines orthogonal to the margin (i.e. EW0113-5, EW0113-6, and repro-n303), SF1 locally 343

contains ≤40 km wide, ≤4.5 km thick wedges of coherent, high-amplitude, dipping 344

reflections that predominantly diverge seaward (Figs 5 and 6); adjacent to the Sonne Ridge 345

on its NW side, a package of dipping reflections diverge landwards (e.g., Fig. 5D). There is 346

no correlation between the location and width of these wedges relative to the magnetic 347

chrons; e.g., some packages of seaward-diverging reflections span several chrons (Fig. 5). 348

Where well-developed wedges are absent, SF1 contains discontinuous, horizontal to gently 349

15

seaward-dipping reflections (Fig. 5). On line s135-11, which is oriented parallel to the 350

margin, most reflections within SF1 are either sub-horizontal or dip gently north-eastwards 351

(Fig. 7). Seismic velocities for SF1 are estimated to be ~4–5 km/s (Fig. 5; Tischer 2006). 352

In places, the uppermost crystalline layer (SF1) is underlain by a low-amplitude, near 353

transparent seismic facies (i.e. SF2), which is particularly clear on lines EW0113-5 and 354

EW0113-6 (Figs 5 and 6). SF2 is up to ~2.8 km thick, being thinnest and occasionally absent 355

beneath wedges of dipping reflections within SF1 (Figs 5 and 6). The few reflections that 356

occur within SF2 typically have low-to-moderate to amplitudes and variable dips (Figs 5 and 357

6). On repro-n303, at the seaward termination of an overlying wedge in SF1, a ~15 km wide 358

swarm of landward-dipping reflections are present in SF2 (Fig. 5D). There is no clear SF2 359

observed on line s135-11, even in areas where it is encountered on the intersecting margin-360

orthogonal lines (Fig. 7). 361

Beneath SF2 we recognise a ~3.5–6 km (<2 s TWT) thick, low-amplitude layer that 362

locally contains prominent, high-amplitude, dipping reflections and discontinuous, moderate 363

amplitude, sub-horizontal reflections (SF3; Figs 5–7). On line EW0113-5, the inclined 364

reflections within SF3 terminate at the Moho and primarily dip oceanwards at 20–30° (Fig. 365

5A). On lines EW0113-6 and repro-n303, however, reflections within SF3 dip both 366

oceanwards and landwards (Figs 5B, C, and 6). Mapped reflections within SF3 on s135-11 367

primarily dip towards the SE, extending from the top of the layer down into the mantle, 368

cross-cutting but not offseting NE-dipping, gently inclined reflections (Fig. 7). Similar mid- 369

and lower-crustal reflection configurations to SF2 and SF3, respectively, occur in the seismic 370

data presented by Hopper et al. (1992) (Fig. 1C). Seismic velocities of SF2 and SF3 are 6.8–371

7.2 km/s (Fig. 4; Tischer 2006). 372

373

Wallaby Plateau and Wallaby Saddle 374

16

Building on previous investigations of seismic data across the continental-to-COTZ crust of 375

the Wallaby Plateau and Wallaby Saddle, here we (re)interpret several 2D seismic lines and 376

compare their structure to that of the CAP. Similar to the CAP, a prominent, continuous, 377

high-amplitude seismic reflection marks the interface between crystalline rock and overlying 378

sedimentary strata across the Wallaby Plateau and Wallaby Saddle (Figs 7 and 8). Within the 379

Wallaby Saddle, the crust appears to be ~5–6 s TWT thick, although the Moho can only 380

tentatively be interpreted, and can also be sub-divided into: (i) SF1, itself containing up to ~4 381

s TWT thick, 12 km wide wedges of diverging reflections that typically dip seawards; (ii) 382

restricted zones that seismically appear similar to SF2 described from the CAP; and (iii) a 383

1.5–3 s TWT thick SF3 unit that contains reflections with variable dips, including prominent 384

swarms of landward-dipping reflections that cross-cut but do not offset other reflections and 385

that typically occur at the oceanward termination of SF1 wedges (Fig. 8). Derivation of 386

interval velocities from seismic reflection stacking velocities suggests rocks comprising SF1 387

have velocities of ~2.5–5.3 km s-1 (see insets in Fig. 8C) (Goncharov & Nelson 2012). It is 388

difficult to determine whether SF1-SF3 continue across the full extent of the Wallaby Saddle 389

in s310-59 because there appears to be a distinct change in reflection configuration (Fig. 8C). 390

In particular, we observe that although reflectivity in west of this change is decreased, 391

reflections towards the top of the crust are broadly sub-parallel to the basement reflection and 392

those within the mid- to lower-crustal areas are either gently inclined landwards, or 393

moderately inclined oceanwards (Fig. 8C). 394

Seismic reflection imaging of the Wallaby Plateau reveals the crust is up to ~7 s TWT 395

thick (e.g., at the Sonne Ridge), thicker than that of the Wallaby Saddle (~5–6 s TWT thick) 396

but that there is no apparent significant change in Moho depth between the two crustal 397

domains (Figs 7 and 8); we note these observations are based on time-migrated data and may 398

thus be invalidated if there any differences in velocity structure between the two areas not 399

17

previously recognised. The crust of the Wallaby Plateau is also thicker than that of the CAP, 400

and its underlying Moho is located at deeper levels (~12 s TWT; Fig. 7). Towards the SW 401

margin of the Wallaby Plateau, a ~40 km wide, apparently NE-trending rift system occurs, 402

comprising normal faults with throws of up to ~1 s TWT that bound and dissect a graben 403

(Figs 8B–D). Reflections within the upper section of the Wallaby Plateau crust are typically 404

moderate-to-high amplitude and form layered packages, which are either conformable to the 405

top basement horizon or that diverge (Fig. 8). The diverging packages of dipping reflections 406

appear similar to SF1 observed in the CAP and Wallaby Saddle (Figs 5 and 8). Derivation of 407

interval velocities from seismic reflection stacking velocities suggests rocks comprising these 408

diverging reflector sequences have velocities of ~2.5–5.3 km s-1 (Fig. 8C) (Goncharov & 409

Nelson 2012). Due to uncertainties regarding the reliability of seismic processing within the 410

middle and lower crustal sections of the Wallaby Plateau, e.g., where imaging is hindered by 411

seabed multiples, it is difficult to confidently interpret reflections as real geological features 412

and not artefacts. However, we note that in these middle and lower crustal sections, 413

reflections are low-to-moderate amplitude and broadly dip gently in various directions; in 414

places, steeply inclined reflections are observed that appear to cross-cut but not offset gently 415

dipping reflections (Fig. 8). These steeply inclined mid- to lower-crustal reflections typically 416

appear to be located beneath diverging reflection packages, or beyond their down-dip 417

termination (Fig. 8). 418

419

Comparison of magnetic anomalies to seismic reflection data 420

EMAG2v2 and ship-track magnetic data reveal that 10 km wide, 220 km long magnetic 421

stripes cover much of the CAP (Figs 1B, 4, and 5). No magnetic stripes can confidently be 422

identified and dated within the Wallaby Plateau and none are observed within the Wallaby 423

Saddle (Figs 1B and 4). Although magnetic anomalies in the EMAG2v3 grid are suppressed 424

18

relative to EMAG2v2, subtle, linear anomalies can still be distinguished across the CAP and 425

in the Gallah Province (cf. Figs 4A and B). Due to the lower resolution of magnetic 426

anomalies in the EMAG2v3 grid, magnetic chrons cannot be confidently attributed and we 427

thus rely on EMAG2v2 and shiptrack data to define possible chrons (Fig. 4). Proximal to the 428

Australian continent, long-wavelength magnetic anomalies can only be broadly assigned to 429

chron M10N (~135.9–134.2 Ma; Figs 1B, 4, and 5) (Robb et al. 2005); across parts of 430

seismic lines EW0113-5, EW0113-6, and repro-n303, chrons M10n–M5r (~135.3–131.4 Ma) 431

are clearly defined and have amplitudes of ≤±100 nT (Figs 1B, 4, and 5). On all three seismic 432

lines, chrons M8r–M7n (~133–132 Ma) coincide with a package of seaward-dipping 433

reflectors observed in SF1, which on EW0113-5 is ≤3 km thick and ~25 km long (Fig. 5). 434

Robb et al. (2005) interpret the magnetic anomalies M10N–M6 southeast of the 435

Sonne Ridge as conjugate to a more poorly developed set of anomalies northwest of the ridge 436

(Fig. 1B). These chrons NW of the Sonne Ridge (i.e. M10N–M6; ~135.9–131.7 Ma) 437

terminate abruptly north-eastwards against the Cape Range Fracture Zone, abutting magnetic 438

chrons on the Gallah Province, and to the SW are cross-cut by chrons (M5n?; 131.7–130.6 439

Ma) mirrored either side of the Sonja Ridge (Fig. 1B). Beyond the outermost chron (M10N) 440

interpreted in the CAP, chron M5n is the first to occur continuously along-strike across both 441

the Cuvier and Gascoyne margin segments, extending into the Wallaby Plateau (Fig. 1B). 442

443

Geochemistry of basalts dredged from the Sonne Ridge 444

The only basalt collected from the Sonne Ridge displays a relatively flat Rare Earth Element 445

(REE) pattern (Fig. 9A). Based on this observation, Dadd et al. (2015) interpret the basalt to 446

have a slightly enriched MORB-like source, supporting the inference that the CAP comprises 447

oceanic crust (e.g., Larson et al. 1979; Hopper et al. 1992; Mihut & Müller 1998). Although a 448

flat REE pattern can be indicative of a shallow melting regime related to MORB generation, 449

19

it does not preclude other settings. By replotting the trace element and radiogenic isotopic 450

compositions of the Sonne Ridge sample, we show the sample has characteristics that could 451

suggest a more continental source (Fig. 9A). It should be noted that the Sonne Ridge sample 452

is heavily altered (Dadd et al. 2015) which could explain the elemental enrichment in Pb, Ba, 453

and Rb, as well as elevated 87Sr/86Sr. However, the sample exhibits unradiogenic εNd and a 454

negative Nb anomaly that is in part defined by a relative enrichment in the neighbouring 455

element Th, which likely cannot be attributed to contamination; the combination of these 456

features likely cannot be ascribed to alteration (Fig. 9). Instead, the negative Nb anomaly and 457

unradiogenic εNd may indicate a chemically evolved, continental or sedimentary contribution 458

to the magmas. The chemical similarity of the Sonne Ridge basalt to two ~124 Ma samples 459

from the Wallaby Plateau (Fig. 9), which is interpreted to comprise intruded continental crust 460

(Daniell et al. 2009; Stilwell et al. 2012; Olierook et al. 2015), is also consistent with a 461

continental or sedimentary contribution to the Sonne Ridge magmas. 462

463

Interpretation and Discussion 464

Since the recognition that it contains magnetic stripes, the CAP has been considered to 465

comprise unambiguous oceanic crust that formed at ~136 Ma (Valanginian) in response to 466

seafloor spreading at the Sonne Ridge (e.g., Falvey & Veevers 1974; Larson et al. 1979; 467

Hopper et al. 1992; Robb et al. 2005). An oceanic origin for the CAP has been supported by 468

seismic reflection-based observations that it has a thin crust relative to adjacent continental 469

blocks (e.g., Fig. 1C) (e.g., Hopper et al. 1992), and chemical data, which suggest it has a 470

MORB-like signature (Dadd et al. 2015). The apparent certainty that the CAP is oceanic 471

means it has been unquestionably treated as such in all geological models of the evolution of 472

NW Australia, including regional and global palinspastic and plate kinematic reconstructions 473

(Heine & Müller 2005; Gibbons et al. 2012). However, the identification of linear magnetic 474

20

anomalies within non-oceanic crust, in areas such as Ethiopia and the Atlantic margins, 475

requires us to re-evaluate the origin of crustal domains previously classified as oceanic crust 476

based on the presence of magnetic stripes (e.g., Bronner et al. 2011; Bridges et al. 2012; 477

Collier et al. 2017; McDermott et al. 2018). 478

479

Seismic facies interpretation 480

Beneath the sedimentary cover across the CAP, we recognise three distinct layers (SF1–SF3; 481

Figs 5–8). We particularly identify a newly recognised upper-crustal layer (SF1) in the CAP 482

that comprises well-developed wedges of divergent, seaward-dipping reflectors (SDRs) (Figs 483

5 and 6). These SDRs are 4.5 km thick, likely have seismic velocities of ~4–5 km s-1 484

(Tischer 2006), and that collectively extend >300 km seaward from the previously interpreted 485

COTZ (Figs 5 and 6). Diverging SDRs are also observed within the: (i) the previously 486

defined, 50–70 km wide COTZ along the Cuvier Margin beneath the continental slope, where 487

they are up to ~5 km thick (e.g., Fig. 1C) (e.g., Hopper et al. 1992; Symonds et al. 1998); and 488

(ii) across the Wallaby Saddle and Wallaby Plateau, where they are ~5–10 km thick and have 489

similar seismic velocities (2.5–5.3 km s-1) to those in the CAP (Figs 5, 6, and 8) (e.g., 490

Symonds et al. 1998; Sayers et al. 2002; Goncharov & Nelson 2012). The lack of boreholes 491

penetrating these SDR sequences offshore NW Australia means we cannot determine their 492

composition or the nature of underlying crust. However, SDR sequences that are 493

geometrically and geophysically (e.g., seismic velocities typically range from ~3–5 km s-1) 494

similar to those from offshore NW Australia (e.g., SF1) have been recognised along other 495

passive margins, developed on both heavily intruded continental crust and thickened oceanic 496

crust (e.g., Hinz 1981; Larsen & Saunders 1998; Harkin et al. 2020). Where these SDRs have 497

been drilled, or are exposed onshore (e.g., Iceland and Greenland), they comprise interbedded 498

basaltic lavas, tuffs, and sedimentary rocks formed during sub-aerial, or perhaps shallow-499

21

water, continental breakup and crustal spreading (e.g., Bodvarsson & Walker 1964; Mutter et 500

al. 1982; Roberts et al. 1984; Eldholm et al. 1987; Larsen et al. 1994a; Geoffroy et al. 2001; 501

Harkin et al. 2020). Based on similarities in structure and seismic velocities to SDRs studied 502

elsewhere, we suggest that SF1 comprises spreading-related volcanic rocks interbedded with 503

sedimentary layers (Figs 1C, 5, 6, and 8) (e.g., Mutter et al. 1982; Hopper et al. 1992; 504

Symonds et al. 1998; Planke et al. 2000; McDermott et al. 2019; Harkin et al. 2020). 505

SDR sequences (SF1) can develop on heavily intruded continental crust or thickened 506

oceanic crust (e.g., Larsen & Saunders 1998). The observed structure and seismic velocities 507

(6.8–7.2 km s-1) of SF2 and SF3 in the CAP, defined by transparent seismic facies and 508

discordant high-amplitude reflections, respectively (Figs 5–8), are consistent with the typical 509

seismic character of sheeted dykes and lower crustal gabbro intrusions in oceanic crust (e.g., 510

Eittreim et al. 1994; Paton et al. 2017). However, we note that these seismic facies are not 511

uniquely diagnostic of oceanic crust but can also occur in COTZs, where moderate- to high-512

amplitude reflections may represent igneous intrusions (e.g., dykes), primary layering within 513

gabbros, or texturally distinct lower crustal shear zones within otherwise homogenous 514

crystalline rocks (e.g., Phipps-Morgan & Chen 1993; Abdelmalak et al. 2015; Paton et al. 515

2017). For example, the swarm of landward dipping reflections within SF2 and SF3 at the 516

down-dip termination of an SDR sequence may correspond to dykes; i.e. they cross-cut but 517

do not offset background reflections and are thus not faults or shear zones (e.g., Figs 5 and 8) 518

(e.g., Abdelmalak et al., 2015; Phillips et al., 2018). 519

520

Implications of SDR recognition for the CAP 521

Origin of SDR lavas 522

Lavas within SDR wedges are inferred to emanate from and be thickest at axial magmatic 523

segments, where they were likely fed by sub-vertical dykes. With continued plate divergence, 524

22

these lavas subside and rotate to dip inwards towards their eruption site (e.g., Planke & 525

Eldholm 1994; Paton et al. 2017; Norcliffe et al. 2018; Tian & Buck 2019); this subsidence 526

also rotates underlying feeder dykes, which will dip away from the magmatic segment (e.g., 527

Lenoir et al. 2003; Abdelmalak et al. 2015). SDRs across the CAP appear to dip and diverge 528

north-westwards, except one SDR-like package of concave-upwards reflections that borders 529

and diverges south-eastwards towards the Sonne Ridge; i.e. we define a conjugate set of 530

SDRs that occur either side of and dip towards the Sonne Ridge (Fig. 5). Although only one 531

SDR package to the NW of the Sonne Ridge dips south-eastwards towards the ridge, we 532

suggest that the other SDR packages, which dip north-westwards, relate to and were formed 533

at the Sonja Ridge (Fig. 5). From these SDR geometries and distribution, coupled with the 534

previously inferred conjugate sets of magnetic chrons (Fig. 1B), our results are consistent 535

with suggestions that (Falvey & Veevers 1974; Larson et al. 1979; Robb et al. 2005; 536

MacLeod et al. 2017): (i) extension within the CAP was predominantly centred on the Sonne 537

Ridge during chrons M10N–M5r (~136–131 Ma); before (ii) briefly jumping to the Sonja 538

Ridge at ~131 Ma (chron M5n), which interrupted subsidence and rotation of the SDR wedge 539

immediately to the NW of the Sonne Ridge and instead produced north-westwards diverging 540

SDRs. There are no changes in SDR divergence direction either side of chron M9n in 541

EW0113-5 or EW0113-6, suggesting no spreading centre existed here (Figs 5A and B) (cf. 542

Gibbons et al. 2012). 543

The chemistry of a basalt sample from the Sonne Ridge, particularly its Nd isotopic 544

composition and refractory trace element abundances (Fig. 9), indicates it could originate 545

from either: (i) melting of sub-continental lithospheric mantle (SCLM); or (ii) contamination 546

of a MORB-like magma by assimilation as it ascended through continental crust (cf. Dadd et 547

al. 2015). Because these chemical data provide evidence that the Sonne Ridge basalt 548

interacted with continental material, these same data cannot thus be used as definitive 549

23

evidence that the CAP comprises oceanic crust (cf. Dadd et al. 2015). The basalt sample 550

dredged from its present-day bathymetric expression can be considered a product of one of 551

the youngest magmas within the CAP system (e.g., Robb et al. 2005). We thus think it 552

plausible that older magmas emanating from the Sonne Ridge, including the SDRs imaged in 553

the seismic reflection data, could have a more pronounced continental signature. 554

555

Environment of SF1 lava emplacement 556

Borehole and field data reveal SDR lavas typically erupt sub-aerially, but can develop sub-557

aqueously (e.g., Bodvarsson & Walker 1964; Mutter et al. 1982; Roberts et al. 1984; Eldholm 558

et al. 1987; Larsen et al. 1994b; Symonds et al. 1998; Planke et al. 2000; Geoffroy et al. 559

2001; Harkin et al. 2020). Determining the environment and age of SDR deposition can help 560

establish whether they likely formed via: (i) seafloor spreading at a mid-ocean ridge, 561

consistent with previous interpretations that the CAP comprises unambiguous oceanic crust 562

(e.g., Falvey & Veevers 1974; Larson et al. 1979; Hopper et al. 1992; Robb et al. 2005); or 563

(ii) magmatic addition along a sub-aerial or shallow-water axis during the transition from 564

continental rifting to full plate separation (i.e. the CAP does not comprise oceanic crust) (e.g., 565

McDermott et al. 2018). However, from their seismic character alone it can be difficult to 566

determine whether SDRs formed in sub-aerial, shallow-water, or deep-marine environments 567

(e.g., compare inner and outer SDR character and inferred emplacement conditions; Symonds 568

et al. 1998; Planke et al. 2000). 569

Observations from the DSDP Site 263 borehole, which terminates ~100–200 m above 570

the crystalline crust, indicate the sedimentary cover deposited above the SDRs: (i) comprises 571

poorly sorted silty claystones that include angular quartz grains and abundant kaolinite, 572

consistent with a neritic (i.e. <200 m water depth) depositional environment (Fig. 3B) (e.g., 573

Robinson et al. 1974; Veevers & Johnstone 1974; Compton et al. 1992; Holbourn & 574

24

Kaminski 1995; Oosting et al. 2006); (ii) contain coarsely agglutinated foraminifera species 575

and taxa such as Hyperamina spp. within the lowermost intersected strata, which are typical 576

of shallow-marine conditions (Holbourn & Kaminski 1995); and (iii) based on 577

biostratigraphic data were deposited at least in the middle Barremian (e.g., ~127 Ma), but are 578

perhaps as old as Hauterivian (~132.6–129.4 Ma) (Oosting et al. 2006). Deposition of the 579

lowermost sedimentary cover intersected by DSDP Site 263 thus occurred up to ~9 Myr (i.e. 580

~135.9–127 Ma) after formation of the CAP crust they rest upon, which records chron M10N 581

(135.9–134.2 Ma), during development of crust hosting chrons M7–M1n (132.5–126.3 Ma; 582

Fig. 1B). Critically, SDR-bearing crust cools and subsides as it is transported away from its 583

emplacement site (e.g., Planke & Eldholm 1994; Paton et al. 2017; Norcliffe et al. 2018; Tian 584

& Buck 2019). The presence of strata deposited in the neritic zone (<200 m water depth) 585

above SF1 in DSDP 263, after ~9 Myr of crustal cooling and subsidence, thus implies lava 586

eruption during the early stages of CAP formation occurred: (i) in a sub-aerial or shallow-587

water environment (i.e. comparable to the inner SDRs of Symonds et al. 1998; Planke et al. 588

2000), if we assume the underlying crust only subsided in the ~9 Myr between SDR 589

emplacement and sediment deposition; or (ii) at a moderately deep-marine spreading centre 590

(i.e. comparable to the outer SDRs of Symonds et al. 1998; Planke et al. 2000), but localised 591

uplift elevated the DSDP 263 area to bathymetric depths equivalent to the neritic zone prior 592

to deposition of overlying strata. We lack the data from strata directly overlying or 593

interbedded with the SDRs to test these two interpretations regarding lava emplacement 594

depth, but note that the relatively flat-lying crystalline crust across the interpreted CAP 595

seismic lines (except for the Sonne Ridge) provides no evidence of post-spreading uplift, 596

perhaps suggesting a sub-aerial or shallow-water environment of emplacement is most 597

plausible (Fig. 5). 598

599

25

Nature of CAP crust 600

Seismic and magnetic data alone are insufficient to determine the origin of the CAP crust 601

because the SDRs, seismic facies (SF1–SF3), and magnetic stripes these data illuminate can 602

be recognised in both oceanic crust and COTZs (e.g., Larsen & Saunders 1998; Symonds et 603

al. 1998; Planke et al. 2000; Bridges et al. 2012; Collier et al. 2017; McDermott et al. 2018). 604

We also show that the chemical data available for a basalt at the Sonne Ridge possesses a 605

continental signature and are thus inconclusive regarding whether or not the crust is oceanic 606

(Fig. 9) (cf. Dadd et al. 2015). However, based on lithological and biostratigraphic data from 607

the sedimentary cover intersected by DSDP Site 263, we suggest: (i) the inferred lavas within 608

SF1, at least during the early stages of CAP formation (i.e. chron M10N), likely erupted in a 609

sub-aerial, or perhaps shallow-water (<200 m water depth), environment; and (ii), assuming 610

the underlying crystalline crust had since subsided relative its position during formation, that 611

the syn-depositional, ~9 Myr old Sonne Ridge was elevated above at least the base of the 612

neritic zone. These constraints on SDR emplacement depth are inconsistent with the CAP 613

being oceanic crust since mid-ocean ridges in such a setting are expected to occur at water 614

depths of ~3 km after 5–10 Myr of spreading (e.g., Menard 1969; Parsons & Sclater 1977; 615

Stein & Stein 1992). 616

We envisage that crustal structure of the CAP could involve a gradual north-617

westwards change from the continental crust of the Cuvier Margin into a COTZ, which is 618

likely characterised at its landward edge by heavily intruded continental crust and 619

progressively becomes increasingly magma-dominated towards the Sonne Ridge (Figs 2 and 620

10). Our data are insufficient to determine where, or if, there is a transition from heavily 621

intruded continental crust to magmatic crust, which would mark break-up of the continental 622

crust within the CAP. Repetition of the M10N-M6 chrons centred on the Sonne Ridge 623

suggests the possible COTZ of the CAP may extend at least out to chron M5n, which is: (i) 624

26

>500 km oceanwards of the outer- limit of the previously defined Cuvier COTZ (e.g., Hopper 625

et al. 1992; Symonds et al. 1998); and (ii) broadly coincident with the north-western limit of 626

the Gallah Province on the Gascoyne margin (Direen et al. 2008) (Figs 1B and 10B). From 627

the distribution of the magnetic chrons (Fig. 1B), our recognition of a continental chemical 628

signature within the Sonne Ridge basalt (Fig. 9), and the probable sub-aerial or shallow-water 629

elevation of the ridge during extension, make it plausible that full continental lithospheric 630

rupture may not have occurred in the CAP (Fig. 10). Instead, we suggest rupture of the 631

continental lithosphere and onset of seafloor spreading occurred simultaneously offshore the 632

Cuvier and Gascoyne margins at ~131 Ma (Hauterivian), following an oceanwards ridge 633

jump from the Sonja Ridge, producing unambiguous oceanic crust recording chron M5 (Figs 634

1 and 10) (e.g., Robb et al. 2005; Direen et al. 2008). Continuation of the COTZ across the 635

CAP has implications for the timing and kinematics of plate reconstructions of the NW 636

Australian margin, with the onset of deep-marine seafloor spreading potentially ~3 Myr later 637

than suggested by previous studies (e.g., Robb et al. 2005). 638

Interpreting the CAP as a COTZ developed through sub-aerial, or at least shallow-639

water, extension implies its crust was: (i) thicker during SDR emplacement, but concurrently 640

and/or subsequently thinned during continued magmatic extension and late-stage stretching 641

(e.g., Bastow & Keir 2011; Bastow et al. 2018); (ii) less dense and thus more buoyant than 642

100% oceanic crust, because it likely retained a significant proportion of continental material; 643

and (iii) thermally buoyant due to the presence of abundant hot intrusions and underlying, 644

decompressing mantle. That these processes can maintain rift axes at above or near sea-level 645

elevations is demonstrated by the onshore occurrence of active rift zones, characterised by 646

heavily-intruded continental crust, in the Main Ethiopia Rift and Afar (e.g., Hayward & 647

Ebinger 1996; Ebinger & Casey 2001; Mackenzie et al. 2005; Bridges et al. 2012). 648

27

Because the degree of thermal subsidence is at least partly controlled by crustal 649

density, we would expect oceanic crust to thermally subside more than less dense, heavily-650

intruded continental crust. Given the Hauterivian-to-Middle Barremian sedimentary strata 651

overlying the SDRs were deposited in neritic conditions (Veevers & Johnstone 1974; 652

Holbourn & Kaminski 1995; Oosting et al. 2006), it is apparent the CAP subsided from near 653

sea-level to a current, unloaded basement depth of ~6.5 km; this total subsidence is greater 654

than predicted for dense, thermally subsiding oceanic crust (Stein & Stein 1992). To interpret 655

the CAP as COTZ crust, our results would require other mechanisms, in addition to thermal 656

subsidence, to influence its subsidence history. For example, post-breakup decay of 657

asthenospheric thermal anomalies may account for some elevation discrepancies via removal 658

of dynamic support of the margin (e.g., Czarnota et al. 2013). Finally, the CAP COTZ may 659

have involved some late-stage stretching prior to terminal breakup and the onset of seafloor 660

spreading, akin to processes observed today in the sub-aerial Red Sea rift in Ethiopia (e.g., 661

Bastow & Keir 2011; Daniels et al. 2014). 662

663

Development of magnetic stripes during break-up 664

Recent forward modelling of conjugate, ship-track magnetic profiles by Collier et al. (2017) 665

suggest magnetic signals over SDRs arise from a combination of stacked and rotated lavas, 666

producing a long-wavelength positive anomaly that can sometimes mask reversals, and linear 667

magnetic anomalies caused by dyke intrusion in the underlying crust. Stacked SDR wedges 668

on the CAP are part of a possible COTZ and span several chrons (e.g. M8n-M7r), but are 669

≤4.5 km thick (Figs 5 and 6). These observations indicate the CAP magnetic stripes likely 670

record magnetic reversal signatures originating from sub-SDR rocks; i.e. the SDRs and flat-671

lying lavas are too thin to dominate the magnetic signature (cf. Collier et al. 2017). In 672

contrast¸ the less-clearly developed yet higher amplitude magnetic reversals in the Gallah 673

28

Province COTZ may relate to interference from the greater SDR thicknesses (≤5.5 km) 674

relative to the CAP (Direen et al. 2008). Our inference that the magnetic signature is derived 675

from sub-SDR rocks is consistent with studies of onshore incipient spreading centres (e.g. 676

Ethiopia), where magnetic stripes likely originate from axial intrusion by dykes in heavily 677

intruded, upper continental crust, rather than overlying lavas (Bridges et al. 2012). 678

We suggest that SDR thickness and, thereby, preservation of magnetic anomalies 679

within a COTZ can partly be attributed to extension rate. For example, the extension rate 680

during SDR eruption offshore NW Australia (~4.5 cm/yr half rate; Robb et al. 2005) is 681

substantially faster than the inferred extension rates for the South Atlantic during magmatic 682

crust formation (~1.1 cm/yr half-rate; Paton et al. 2017). Slower extension rates (e.g. South 683

Atlantic) likely promote stacking of lava flows to produce thicker SDRs (Eagles et al. 2015), 684

leading to interference between the magnetic signal of the SDRs and sub-SDR crust and thus 685

the development of the long-wavelength positive magnetic anomalies (e.g., Moulin et al. 686

2010). Extension rate may also influence magnetic anomaly development by affecting the 687

width of magnetic stripes; reversal anomalies will be narrowest at slow spreading ridges 688

(Vine 1966). The narrower anomalies, combined with the greater potential for vertical 689

stacking of lavas, will tend to suppress magnetic anomaly preservation. 690

691

Conclusions 692

The recognition of magnetic stripes within the Cuvier Abyssal Plain (CAP), offshore NW 693

Australia, has led to the assumption that it comprises oceanic crust generated by conventional 694

seafloor spreading at the Sonne Ridge, probably at water depths of ≳2 km. We challenge this 695

assumption, in line with the growing consensus that magnetic stripes are not necessarily 696

diagnostic of oceanic crust and can instead form in continent-ocean transition zones 697

(COTZs). Using regional 2D seismic reflection lines we demonstrate that the uppermost layer 698

29

in the CAP crystalline line crust contains seaward-dipping reflector (SDR) sequences, akin to 699

those observed in the previously defined COTZ of the Cuvier Margin and Wallaby Saddle, as 700

well as on the heavily intruded continental crust of the Wallaby Plateau. Through comparison 701

to SDRs recognised elsewhere, we suggest those observed across the CAP comprise lavas, 702

interbedded with sedimentary strata, erupted from an axial magmatic segment. Lithological 703

and biostratigraphic data from a borehole penetrating the CAP sedimentary cover, which 704

were deposited in neritic (<200 m water depth) conditions, require the underlying crystalline 705

crust to have been at shallow-water depths ~9 Myr after its formation and thus imply SDR 706

emplacement occurred in a shallow water or sub-aerial environment. We also re-interpret 707

chemical data from a basalt dredged along the Sonne Ridge and, contrary to previous work, 708

show that it exhibits a continental chemical signature. Overall, these data and interpretations 709

suggest the CAP may not comprise unambiguous oceanic crust, but could instead represent a 710

>500 km wide COTZ where extension likely became more magma-dominated, producing 711

heavily-intruded continental crust (akin to present-day Ethiopia) through to magmatic crust. 712

In our model, break-up of the continental crust occurred during the formation of the CAP, but 713

full continental lithospheric rupture occurred outboard of the COTZ following a ridge jump at 714

~130 Ma. Our re-evaluation of the CAP crustal type supports suggestions that COTZs along 715

volcanic passive margins may record the development of magnetic stripes, which thus should 716

not be used alone as a reliable proxy for the onset of seafloor spreading and the extent of 717

oceanic crust. 718

719

Acknowledgements 720

Schlumberger are thanked for provision of Petrel software licenses. M.T.R. was supported by 721

NERC grant NE/L501621/L. The EW0113 seismic survey and EMAG2 magnetic anomaly 722

grids were, and can be acquired from, the UTIG Marine Geoscience Data System and the 723

30

NOAA National Geophysical Data Centre, respectively. Gwenn Peron-Pinvidic, Gareth 724

Roberts, and Saskia Goes are thanked for helpful discussions during the preparation of this 725

manuscript. We thank five reviewers, including Jon Bull, for their constructive comments on 726

previous versions of this manuscript. 727

728

Figure captions 729

Figure 1: (A) Location map of the study area highlighting the seismic lines used in this study 730

and key tectonic elements, including areas of recognised seaward-dipping reflectors (SDRs) 731

(Symonds et al. 1998; Holford et al. 2013) and previously interpreted approximate limits of 732

the continent-ocean boundary (COB; Eagles et al. 2015) and continent-ocean transition zones 733

(COTZs; Symonds et al. 1998; Direen et al. 2008). Inset: study area location offshore NW 734

Australia. AAP – Argo Abyssal Plain, CAP – Cuvier Abyssal Plain, CRFZ – Cape Range 735

Fracture Zone, GAP – Gascoyne Abyssal Plain, GP – Gallah Province, NCB – North 736

Carnarvon Basin, EP – Exmouth Plateau, PB – Perth Basin, SCB – South Carnarvon Basin, 737

Cu – Cuvier margin COTZ, SR – Sonne Ridge, SjR – Sonja Ridge, WP – Wallaby Plateau, 738

WS – Wallaby Saddle, WZFZ – Wallaby-Zenith Fracture Zone. Dredge sites 52, 55 (samples 739

055BS004A and 055BS004B), and 57 (sample 057DR051A) are also shown (Dadd et al. 740

2015). (B) Total magnetic intensity grid (EMAG2v2), interpreted magnetic chrons (based on 741

Robb et al. 2005). See Supplementary Figure S1 for an uninterpreted version. (C) 742

Uninterpreted and interpreted seismic line (i.e. seismic profile 670) across the Cuvier Margin, 743

imaging the crustal structure beneath the continental shelf and the deep abyssal plain 744

(modified from Hopper et al., 1992). Velocity profiles from refraction experiments shown; 745

see Hopper et al., (1992) for details. See Figure 1A for approximate line location and 746

Supplementary Figure S2 for an enlarged version of the uninterpreted seismic line. 747

748

31

Figure 2: Schematic model (not to scale) of a continent-ocean transition zone along a magma-749

rich passive margin, which depicts the evolution from unambiguous continental crust to 750

unambiguous oceanic crust; for simplicity the lithospheric mantle is not shown. As magma 751

intrudes continental crust, likely as dykes at mid- to upper-crustal levels and larger gabbroic 752

bodies in the lower crust, it becomes ‘heavily intruded continental crust’ (e.g., Eldholm et al. 753

1989). Continued intrusion and dyking leads to localisation of magmatism within narrow 754

zones where there is little, if any, continental crust remaining (i.e. 'magmatic crust'; e.g., 755

Collier et al. 2017; Paton et al. 2017). We categorize heavily intruded continental crust and 756

magmatic crust as ‘COTZ crust’. Sub-aerial, magma-assisted rifting may feed extensive lava 757

flows that later, through subsidence, become seaward-dipping reflectors (SDRs). SDR 758

subsidence leads to rotation of underlying dykes (Abdelmalak et al. 2015); a similar rotation 759

of lavas and dykes is observed in oceanic crust (Karson 2019). 760

761

Figure 3: Tectono-stratigraphic chart for the Exmouth Plateau and Cuvier Margin (after 762

Hocking et al., 1987; Arditto, 1993; Partington et al. 2003; Willis, 2005; Reeve et al. 2016). 763

(B) Comparison between stratigraphic data from DSDP 263 and Pendock-1 boreholes 764

(modified from Veevers & Johnstone, 1974; Holbourn & Kaminski, 1995). See Figure 1A for 765

borehole locations. 766

767

Figure 4: Total magnetic intensity grids EMAG2v2 and EMAG2v3 (Maus et al. 2009; Meyer 768

et al. 2017), compared with shiptrack magnetic data (Robb et al. 2005). Key tectonic 769

elements also shown (see Fig. 1 for legend). In (A), CAP – Cuvier Abyssal Plain, GAP – 770

Gascoyne Abyssal Plain, GP – Gallah Province, NCB – North Carnarvon Basin, PB – Perth 771

Basin, SCB – South Carnarvon Basin, Cu – Cuvier margin COTZ, WP – Wallaby Plateau, 772

WS – Wallaby Saddle. 773

32

774

Figure 5: Interpreted and uninterpreted, depth-converted seismic lines (A) EW0113-5 and (B) 775

EW0113-6, and the time-migrated line (D) repro n303 showing crustal structure of the Cuvier 776

Margin; see Figures 1A and 5C for line locations. The tie-co-located magnetic anomaly 777

profile showing interpreted magnetic chrons is presented for (A–D) (after Robb et al. 2005). 778

See Supplementary Figure S2 for an enlarged version of the uninterpreted seismic lines. 779

780

Figure 6: Zoomed in view of EW0113-5 highlighting the seismic character of interpreted 781

SDR packages (see Fig. 5A for location). 782

783

Figure 7: Interpreted and uninterpreted, time-migrated seismic line s135-11; see Figure 1A 784

for location. See Supplementary Figure S2 for an enlarged version of the uninterpreted 785

seismic line. 786

787

Figure 8: Interpreted and uninterpreted, time-migrated seismic lines (A) s135-s135_05, (B) 788

s135-08, and (D) s310-59 showing crustal structure of the Wallaby Plateau and Wallaby 789

Saddle; see Figures 1A and 8D for line locations. See Supplementary Figure S2 for an 790

enlarged version of the uninterpreted seismic lines. 791

792

Figure 9: (A) Primitive mantle normalized incompatible element diagram comparing the 793

dredged Sonne Ridge and Wallaby Plateau basalt lava samples with average (ave.) 794

compositions of MORB variants (Hofmann 2014), Globally Subducting Sediment (GLOSS) 795

(Plank & Langmuir 1998), and continental crust (Rudnick & Fountain 1995). (B) Plot of 796

ε(Nd) versus 87Sr/86Sr, illustrating that the Sonne Ridge and Wallaby Plateau samples are 797

distinct from MORB (based on data collated in Hofmann 2014). 798

33

799

Figure 10: (A) Map showing the potential limits of the COTZ based on interpreting the CAP 800

and Gallah Province as transitional and/or magmatic crust. (B-D) Schematic maps showing 801

the development of COTZ crust and the onset of oceanic crust accretion adjacent to the 802

Gascoyne and Cuvier margins, during formation of chrons (B) M10, (C) M6 and (D) M3r. 803

See Figure 1 for chron ages. Location of present day coastline shown for reference. 804

805

References 806

Abdelmalak, M.M., Andersen, T.B., Planke, S., Faleide, J.I., Corfu, F., Tegner, C., Shephard, 807 G.E., Zastrozhnov, D., et al. 2015. The ocean-continent transition in the mid-Norwegian 808 margin: Insight from seismic data and an onshore Caledonian field analogue. Geology, 43, 809 1011-1014. 810

811 Bastow, I.D. & Keir, D. 2011. The protracted development of the continent-ocean transition 812 in Afar. Nature Geosci, 4, 248-250. 813

814 Bastow, I.D., Booth, A.D., Corti, G., Keir, D., Magee, C., Jackson, C.A.L., Warren, J., 815 Wilkinson, J., et al. 2018. The Development of Late‐Stage Continental Breakup: Seismic 816 Reflection and Borehole Evidence from the Danakil Depression, Ethiopia. 37, 2848-2862. 817

818 Bodvarsson, G. & Walker, G. 1964. Crustal drift in Iceland. Geophysical Journal 819 International, 8, 285-300. 820

821 Bolli, H.M. 1974. Jurassic and Cretaceous Calcisphaerulidae from DSDP Leg 27, eastern 822 Indian Ocean. 823

824 Bridges, D.L., Mickus, K., Gao, S.S., Abdelsalam, M.G. & Alemu, A. 2012. Magnetic stripes 825 of a transitional continental rift in Afar. Geology, 40, 203-206. 826

827 Bronner, A., Sauter, D., Manatschal, G., Péron-Pinvidic, G. & Munschy, M. 2011. Magmatic 828 breakup as an explanation for magnetic anomalies at magma-poor rifted margins. Nature 829 Geoscience, 4, 549. 830

831 Causer, A., Pérez-Díaz, L., Adam, J. & Eagles, G. 2020. Uncertainties in break-up markers 832 along the Iberia–Newfoundland margins illustrated by new seismic data. Solid Earth, 11, 833 397-417. 834

835

34

Collier, J.S., McDermott, C., Warner, G., Gyori, N., Schnabel, M., McDermott, K. & Horn, 836 B.W. 2017. New constraints on the age and style of continental breakup in the South Atlantic 837 from magnetic anomaly data. Earth and Planetary Science Letters, 477, 27-40. 838

839 Colwell, J., Symonds, P. & Crawford, A. 1994. The nature of the Wallaby (Cuvier) Plateau 840 and other igneous provines of the west Australian margin. Journal of Australian Geology and 841 Geophysics, 15, 137-156. 842

843 Compton, J., Mallinson, D., Netranatawong, T. & Locker, D. 1992. Regional correlation of 844 mineralogy and diagenesis of sediment from the Exmouth Plateau and Argo Basin, 845 Northwestern Australian Continental Margin. 846

847 Corti, G., Agostini, A., Keir, D., Van Wijk, J., Bastow, I.D. & Ranalli, G. 2015. Magma-848 induced axial subsidence during final-stage rifting: Implications for the development of 849 seaward-dipping reflectors. Geosphere, 11, 563-571. 850

851 Crawford, A.J. & von Rad, U. 1994. The petrology, geochemistry and implications of basalts 852 dredged from the Rowley Terrace-Scott Plateau and Exmouth Plateau margins, northwestern 853 Australia. Journal of Australian Geology and Geophysics, 15, 43-54. 854

855 Czarnota, K., Hoggard, M., White, N. & Winterbourne, J. 2013. Spatial and temporal patterns 856 of Cenozoic dynamic topography around Australia. Geochemistry, Geophysics, Geosystems, 857 14, 634-658. 858

859 Dadd, K.A., Kellerson, L., Borissova, I. & Nelson, G. 2015. Multiple sources for volcanic 860 rocks dredged from the Western Australian rifted margin. Marine Geology, 368, 42-57. 861

862 Daniell, J., Jorgensen, D., Anderson, T., Borissova, I., Burq, S., Heap, A., Hughes, D., 863 Mantle, D., et al. 2009. Frontier basins of the West Australian continental margin. 864 Geoscience Australia Record, 38, 243. 865

866 Daniels, K.A., Bastow, I.D., Keir, D., Sparks, R.S.J. & Menand, T. 2014. Thermal models of 867 dyke intrusion during development of continent–ocean transition. Earth and Planetary 868 Science Letters, 385, 145-153. 869

870 Direen, N.G., Stagg, H.M.J., Symonds, P.A. & Colwell, J.B. 2008. Architecture of volcanic 871 rifted margins: new insights from the Exmouth – Gascoyne margin, Western Australia. 872 Australian Journal of Earth Sciences, 55, 341-363. 873

874 Direen, N.G., Borissova, I., Stagg, H., Colwell, J.B. & Symonds, P.A. 2007. Nature of the 875 continent–ocean transition zone along the southern Australian continental margin: a 876 comparison of the Naturaliste Plateau, SW Australia, and the central Great Australian Bight 877 sectors. Geological Society, London, Special Publications, 282, 239-263. 878

879

35

Eagles, G., Pérez-Díaz, L. & Scarselli, N. 2015. Getting over continent ocean boundaries. 880 Earth-Science Reviews, 151, 244-265. 881

882 Ebinger, C.J. & Casey, M. 2001. Continental breakup in magmatic provinces: An Ethiopian 883 example. Geology, 29, 527. 884

885 Eittreim, S.L., Gnibidenko, H., Helsley, C.E., Sliter, R., Mann, D. & Ragozin, N. 1994. 886 Oceanic crustal thickness and seismic character along a central Pacific transect. Journal of 887 Geophysical Research: Solid Earth, 99, 3139-3145. 888

889 Eldholm, O., Thiede, J. & Taylor, E. 1989. The Norwegian continental margin: tectonic, 890 volcanic, and paleoenvironmental framework. Proceedings of the ocean drilling program, 891 Scientific results. Citeseer, 5-26. 892

893 Eldholm, O., Thiede, J., Taylor, E. & Party, S.S. 1987. Summary and preliminary 894 conclusions, ODP Leg 104. Proceedings of the Ocean Drilling Program, Scientific Results. 895 Ocean Drilling Program College Station, Texas, 751-771. 896

897 Falvey, D. & Veevers, J. 1974. Physiography of the Exmouth and Scott plateaus, western 898 Australia, and adjacent northeast Wharton Basin. Marine Geology, 17, 21-59. 899

900 Geoffroy, L., Callot, J.P., Scaillet, S., Skuce, A., Gélard, J., Ravilly, M., Angelier, J., Bonin, 901 B., et al. 2001. Southeast Baffin volcanic margin and the North American‐Greenland plate 902 separation. Tectonics, 20, 566-584. 903

904 Gibbons, A.D., Barckhausen, U., den Bogaard, P., Hoernle, K., Werner, R., Whittaker, J.M. 905 & Müller, R.D. 2012. Constraining the Jurassic extent of Greater India: Tectonic evolution of 906 the West Australian margin. Geochemistry, Geophysics, Geosystems, 13. 907

908 Goncharov, A. & Nelson, G. 2012. From two way time to depth and pressure for 909 interpretation of seismic velocities offshore: Methodology and examples from the Wallaby 910 Plateau on the West Australian margin. Tectonophysics, 572, 26-37. 911

912 Gradstein, F. & Ogg, J. 2012. The chronostratigraphic scale The geologic time scale. 913 Elsevier, 31-42. 914

915 Harkin, C., Kusznir, N., Roberts, A., Manatschal, G. & Horn, B. 2020. Origin, composition 916 and relative timing of seaward dipping reflectors on the Pelotas rifted margin. Marine and 917 petroleum geology, 114, 104235. 918

919 Hayward, N. & Ebinger, C. 1996. Variations in the along‐axis segmentation of the Afar Rift 920 system. Tectonics, 15, 244-257. 921

922

36

Heine, C. & Müller, R. 2005. Late Jurassic rifting along the Australian North West Shelf: 923 margin geometry and spreading ridge configuration. Australian Journal of Earth Sciences, 924 52, 27-39. 925

926 Hey, R. 1977. A new class of “pseudofaults” and their bearing on plate tectonics: A 927 propagating rift model. Earth and Planetary Science Letters, 37, 321-325. 928

929 Hinz, K. 1981. A hypothesis on terrestrial catastrophies Wedges of very thick oceanward 930 dipping layers beneath passive continental margins. Their origin and paleoenvironmental 931 significance. Geologisches Jahrbuch. Reihe E, Geophysik, 3-28. 932

933 Hofmann, A. 2014. Sampling mantle heterogeneity through oceanic basalts: Isotopes and 934 trace elements. In: RW, C. (ed) The Mantle and Core, Treatise on Geochemistry. Elsevier-935 Pergamon, Oxford, 67-101. 936

937 Holbourn, A.E. & Kaminski, M.A. 1995. Lower Cretaceous benthic foraminifera from DSDP 938 Site 263: micropalaeontological constraints for the early evolution of the Indian Ocean. 939 Marine Micropaleontology, 26, 425-460. 940

941 Holford, S.P., Schofield, N., Jackson, C.A.L., Magee, C., Green, P.F. & Duddy, I.R. 2013. 942 Impacts of igneous intrusions on source and reservoir potential in prospective sedimentary 943 basins along the western Australian continental margin. In: Keep, M. & Moss, S.J. (eds) The 944 Sedimentary Basins of Western Australia IV. Proceedings of the Petroleum Exploration 945 Society of Australia Symposium, Perth, WA. 946

947 Hopper, J.R., Mutter, J.C., Larson, R.L. & Mutter, C.Z. 1992. Magmatism and rift margin 948 evolution: Evidence from northwest Australia. Geology, 20, 853-857. 949

950 Karson, J.A. 2019. From Ophiolites to Oceanic Crust: Sheeted Dike Complexes and Seafloor 951 Spreading. In: Srivastava, R., Ernst, R. & Peng, P. (eds) Dyke Swarms of the World: A 952 Modern Perspective. Springer, 459-492. 953

954 Keranen, K., Klemperer, S., Gloaguen, R. & Group, E.W. 2004. Three-dimensional seismic 955 imaging of a protoridge axis in the Main Ethiopian rift. Geology, 32, 949-952. 956

957 Larsen, H. & Saunders, A. 1998. 41. Tectonism and volcanism at the Southeast Greenland 958 rifted margin: a record of plume impact and later continental rupture. Proceedings of the 959 Ocean Drilling Program, Scientific Results, 503-533. 960

961 Larsen, H., Saunders, A. & Clift, P. 1994a. Proceedings of the Ocean Drilling Program, 962 Initial Reports. Ocean Drilling Program, College Station, Texas, 1-152. 963

964

37

Larsen, H., Saunders, A., Larsen, L. & Lykke-Andersen, H. 1994b. ODP activities on the 965 South-East Greenland margin: Leg 152 drilling and continued site surveying. Rapport 966 Grønlands Geologiske Undersøgelse, 160, 75-81. 967

968 Larson, R.L., Mutter, J.C., Diebold, J.B., Carpenter, G.B. & Symonds, P. 1979. Cuvier Basin: 969 a product of ocean crust formation by Early Cretaceous rifting off Western Australia. Earth 970 and Planetary Science Letters, 45, 105-114. 971

972 Lenoir, X., Féraud, G. & Geoffroy, L. 2003. High-rate flexure of the East Greenland volcanic 973 margin: constraints from 40Ar/39Ar dating of basaltic dykes. Earth and Planetary Science 974 Letters, 214, 515-528. 975

976 Longley, I., Buessenschuett, C., Clydsdale, L., Cubitt, C., Davis, R., Johnson, M., Marshall, 977 N., Murray, A., et al. 2002. The North West Shelf of Australia–a Woodside perspective. The 978 sedimentary basins of Western Australia, 3, 27-88. 979

980 Mackenzie, G., Thybo, H. & Maguire, P. 2005. Crustal velocity structure across the Main 981 Ethiopian Rift: results from two-dimensional wide-angle seismic modelling. Geophysical 982 Journal International, 162, 994-1006. 983

984 MacLeod, S.J., Williams, S.E., Matthews, K.J., Müller, R.D. & Qin, X. 2017. A global 985 review and digital database of large-scale extinct spreading centers. Geosphere, 13, 911-949. 986

987 Maguire, P., Keller, G., Klemperer, S., Mackenzie, G., Keranen, K., Harder, S., O Reilly, B., 988 Thybo, H., et al. 2006. Crustal structure of the northern Main Ethiopian Rift from the 989 EAGLE controlled-source survey; a snapshot of incipient lithospheric break-up. Geological 990 Society, London, Special Publications, 259, 269. 991

992 Maus, S., Barckhausen, U., Berkenbosch, H., Bournas, N., Brozena, J., Childers, V., 993 Dostaler, F., Fairhead, J., et al. 2009. EMAG2: A 2–arc min resolution Earth Magnetic 994 Anomaly Grid compiled from satellite, airborne, and marine magnetic measurements. 995 Geochemistry, Geophysics, Geosystems, 10. 996

997 McDermott, C., Lonergan, L., Collier, J.S., McDermott, K.G. & Bellingham, P. 2018. 998 Characterization of Seaward‐Dipping Reflectors Along the South American Atlantic Margin 999 and Implications for Continental Breakup. Tectonics, 37, 3303-3327. 1000

1001 McDermott, C., Collier, J.S., Lonergan, L., Fruehn, J. & Bellingham, P. 2019. Seismic 1002 velocity structure of seaward-dipping reflectors on the South American continental margin. 1003 Earth and Planetary Science Letters, 521, 14-24. 1004

1005 Menard, H. 1969. Elevation and subsidence of oceanic crust. Earth and Planetary Science 1006 Letters, 6, 275-284. 1007

1008

38

Menzies, M., Klemperer, S., Ebinger, C. & Baker, J. 2002. Characteristics of volcanic rifted 1009 margins. In: Menzies, M., Klemperer, S., Ebinger, C. & Baker, J. (eds) Volcanic Rifted 1010 Margins, Special Publications. Geological Society of America, 362, 1-14. 1011

1012 Meyer, B., Chulliat, A. & Saltus, R. 2017. Derivation and error analysis of the Earth 1013 Magnetic Anomaly Grid at 2 arc min Resolution Version 3 (EMAG2v3). Geochemistry, 1014 Geophysics, Geosystems, 18, 4522-4537. 1015

1016 Mihut, D. & Müller, R.D. 1998. Volcanic margin formation and Mesozoic rift propagators in 1017 the Cuvier Abyssal Plain off Western Australia. Journal of Geophysical Research, 103, 1018 27149. 1019

1020 Moulin, M., Aslanian, D. & Unternehr, P. 2010. A new starting point for the South and 1021 Equatorial Atlantic Ocean. Earth-Science Reviews, 98, 1-37. 1022

1023 Mutter, J.C., Talwani, M. & Stoffa, P.L. 1982. Origin of seaward-dipping reflectors in 1024 oceanic crust off the Norwegian margin by “subaerial sea-floor spreading”. Geology, 10, 353-1025 357. 1026

1027 Norcliffe, J.R., Paton, D.A., Mortimer, E.J., McCaig, A.M., Nicholls, H., Rodriguez, K., 1028 Hodgson, N. & Van Der Spuy, D. 2018. Laterally Confined Volcanic Successions (LCVS); 1029 recording rift-jumps during the formation of magma-rich margins. Earth and Planetary 1030 Science Letters, 504, 53-63. 1031

1032 Olierook, H.K., Merle, R.E., Jourdan, F., Sircombe, K., Fraser, G., Timms, N.E., Nelson, G., 1033 Dadd, K.A., et al. 2015. Age and geochemistry of magmatism on the oceanic Wallaby 1034 Plateau and implications for the opening of the Indian Ocean. Geology, 43, 971-974. 1035

1036 Oosting, A., Leereveld, H., Dickens, G., Henderson, R. & Brinkhuis, H. 2006. Correlation of 1037 Barremian-Aptian (mid-Cretaceous) dinoflagellate cyst assemblages between the Tethyan 1038 and Austral realms. Cretaceous Research, 27, 792-813. 1039

1040 Parsons, B. & Sclater, J.G. 1977. An analysis of the variation of ocean floor bathymetry and 1041 heat flow with age. Journal of Geophysical Research, 82, 803-827. 1042

1043 Paton, D., Pindell, J., McDermott, K., Bellingham, P. & Horn, B. 2017. Evolution of 1044 seaward-dipping reflectors at the onset of oceanic crust formation at volcanic passive 1045 margins: Insights from the South Atlantic. Geology, 45, 439-442. 1046

1047 Peron-Pinvidic, G., Manatschal, G. & Participants, a.t.I.R.W. 2019. Rifted margins: state of 1048 the art and future challenges. Frontiers in Earth Science, 7, 8. 1049

1050

39

Phipps-Morgan, J. & Chen, Y.J. 1993. The genesis of oceanic crust: Magma injection, 1051 hydrothermal circulation, and crustal flow. Journal of Geophysical Research: Solid Earth, 1052 98, 6283-6297. 1053

1054 Pickup, S., Whitmarsh, R., Fowler, C. & Reston, T. 1996. Insight into the nature of the 1055 ocean-continent transition off West Iberia from a deep multichannel seismic reflection 1056 profile. Geology, 24, 1079-1082. 1057

1058 Plank, T. & Langmuir, C.H. 1998. The chemical composition of subducting sediment and its 1059 consequences for the crust and mantle. Chemical Geology, 145, 325-394. 1060

1061 Planke, S. & Eldholm, O. 1994. Seismic response and construction of seaward dipping 1062 wedges of flood basalts: Vøring volcanic margin. Journal of Geophysical Research: Solid 1063 Earth, 99, 9263-9278. 1064

1065 Planke, S., Symonds, P.A., Alvestad, E. & Skogseid, J. 2000. Seismic volcanostratigraphy of 1066 large‐volume basaltic extrusive complexes on rifted margins. Journal of Geophysical 1067 Research: Solid Earth, 105, 19335-19351. 1068

1069 Rabinowitz, P.D. & LaBrecque, J. 1979. The Mesozoic South Atlantic Ocean and evolution 1070 of its continental margins. Journal of Geophysical Research: Solid Earth, 84, 5973-6002. 1071

1072 Robb, M.S., Taylor, B. & Goodliffe, A.M. 2005. Re-examination of the magnetic lineations 1073 of the Gascoyne and Cuvier Abyssal Plains, off NW Australia. Geophysical Journal 1074 International, 163, 42-55. 1075

1076 Roberts, D., Backman, J., Morton, A., Murray, J. & Keene, J. 1984. Evolution of volcanic 1077 rifted margins – synthesis of leg-81 results on the West margin of Rockall Plateau. 1078

1079 Robinson, P.T., Thayer, P., Cook, P., McKnight, B. & et al. 1974. Lithology of Mesozoic and 1080 Cenozoic sediments of the eastern Indian Ocean, Leg 27, Deep Sea Drilling Project. 1081

1082 Rooney, T.O., Bastow, I.D., Keir, D., Mazzarini, F., Movsesian, E., Grosfils, E.B., 1083 Zimbelman, J.R., Ramsey, M.S., et al. 2014. The protracted development of focused 1084 magmatic intrusion during continental rifting. Tectonics, 33, 875-897. 1085

1086 Rudnick, R.L. & Fountain, D.M. 1995. Nature and composition of the continental crust: a 1087 lower crustal perspective. Reviews of Geophysics, 33, 267-309. 1088

1089 Sayers, J., Borissova, I., Ramsay, D. & Symonds, P. 2002. Geological framework of the 1090 Wallaby Plateau and adjacent areas. 1091

1092

40

Scheibnerová, V. 1974. Aptian–Albian benthonic foraminifera from DSDP Leg 27, Sites 259, 1093 260 and 263, Eastern Indian Ocean. 1094

1095 Skogseid, J., Pedersen, T., Eldholm, O. & Larsen, B.T. 1992. Tectonism and magmatism 1096 during NE Atlantic continental break-up: the Voring Margin. Geological Society, London, 1097 Special Publications, 68, 305-320. 1098

1099 Skogseid, J., Planke, S., Faleide, J.I., Pedersen, T., Eldholm, O. & Neverdal, F. 2000. NE 1100 Atlantic continental rifting and volcanic margin formation. In: Nottvedt, A. (ed) Dynamics of 1101 the Norwegian Margin. Geological Society, London, Special Publications, London, 167, 295-1102 326. 1103

1104 Stagg, H., Alcock, M., Bernardel, G., Moore, A., Symonds, P. & Exon, N. 2004. Geological 1105 framework of the outer Exmouth Plateau and adjacent ocean basins. Geoscience Australia. 1106

1107 Stein, C.A. & Stein, S. 1992. A model for the global variation in oceanic depth and heat flow 1108 with lithospheric age. Nature, 359, 123. 1109

1110 Stilwell, J., Quilty, P. & Mantle, D. 2012. Paleontology of Early Cretaceous deep-water 1111 samples dredged from the Wallaby Plateau: new perspectives of Gondwana break-up along 1112 the Western Australian margin. Australian Journal of Earth Sciences, 59, 29-49. 1113

1114 Symonds, P.A., Planke, S., Frey, O. & Skogseid, J. 1998. Volcanic evolution of the Western 1115 Australian Continental Margin and its implications for basin development. The Sedimentary 1116 Basins of Western Australia 2: Proc. of Petroleum Society Australia Symposium, Perth, WA. 1117

1118 Talwani, M. & Eldholm, O. 1973. Boundary between continental and oceanic crust at the 1119 margin of rifted continents. Nature, 241, 325. 1120

1121 Tian, X. & Buck, W.R. 2019. Lithospheric thickness of volcanic rifting margins: Constraints 1122 from seaward dipping reflectors. Journal of Geophysical Research: Solid Earth, 124, 3254-1123 3270. 1124

1125 Tischer, M. 2006. The structure and development of the continent-ocean transition zone of 1126 the Exmouth Plateau and Cuvier margin, Northwest Australia: implications for extensional 1127 strain partitioning. PhD, Columbia University. 1128

1129 Veevers, J. 1986. Breakup of Australia and Antarctica estimated as mid-Cretaceous (95±5 1130 Ma) from magnetic and seismic data at the continental margin. Earth and Planetary Science 1131 Letters, 77, 91-99. 1132

1133 Veevers, J. & Johnstone, M. 1974. Comparative stratigraphy and structure of the western 1134 Australian margin and the adjacent deep ocean floor. Initial Reports of the Deep Sea Drilling 1135 Project, 27, 571-585. 1136

41

1137 Vine, F.J. 1966. Spreading of the ocean floor: new evidence. Science, 154, 1405-1415. 1138

1139 Vine, F.J. & Matthews, D.H. 1963. Magnetic anomalies over oceanic ridges. Nature, 199, 1140 947-949. 1141

1142 Wiseman, J.F. & Williams, A. 1974. Palynological investigation of samples from sites 259, 1143 261, and 263, Leg 27, Deep Sea Drilling Project. 1144

1145

1146

1147

1148

1149

1150

1151

1152

1153

1154

1155

1156

1157

1158

1159

1160

1161

1162

1163

1164

1165

1166

42

Table 1: Interval velocities

Layer Seismic velocity (km s-

1)

Water column 1.5

Sedimentary strata 2.0–2.8 Seaward-dipping reflectors (SDRs) 4.9

Sub-SDR crust 6.8–7.2

Upper mantle 8

1167

(C)0

2

4

6

8

10

Two-

way

trav

el-ti

me

(s) W E

20 km

0

2

4

6

8

10

Two-

way

trav

el-ti

me

(s)

20 km

Seabed

Top crystallinebasement

0 8 km s-1

0 8 km s-1

0 8 km s-1

Seabed multiple

Moho

SDRs

Interpreted reflections from Hopper et al., (1992)

Sedimentary strata

Continental crust

108°E 110°E 112°E 114°E

0 100 200 300 400 500 km-300 0 300Magnetic anomaly (nT)

N

M10N

M10M9

M7

M8

M6M6M7M8?

M10

M10N

M5nM5n

M10N

M8

M6

M10M9?

M10M6

M5n

M5n

M1n

M3nM1n

M1nM3n

M5n

(B)

M8?M6?

M10?

M10N M10 M9 M8 M7 M6 M5n M3n M1n

(135.9 - 134.2 Ma) (134.2 - 133.6 Ma) (133.6 - 133.0 Ma) (133.0 - 132.5 Ma) (132.5 - 132.0 Ma) (132.0 - 131.7 Ma) (131.7 - 130.6 Ma) (130.6 - 128.7 Ma) (128.7 - 126.3 Ma)

M5nM3n

108°E 110°E 112°E 114°E

26°S

24°S

22°S

20°S

18°S

16°S

001 -6 100 200 300 400 500 kmN

EW0113-6

EW0113-5DSDP26357

Elevation (km)

(A)

Fig. 1C

CRFZ

AAPGAP

CAP

2000 km

NCBAustralia

NCB

SCB

PB

WP

WZFZ

SRSjR

WS

CAP

GAP

GP

Cu repro-n303s135-135_05

s135-135_08s310-s310_59

s135-135_11

(Fig. 7)

55

GascoyneMargin

CuvierMargin

Pendock-1

52

Tectonic element boundaryCOB (see Eagles et al., 2015)COTZ (Symonds et al., 1998)SDRs (Symonds et al., 1998)Extinct spreading centre (Robb et al., 2005)Extinct spreading centre (Gibbons et al., 2012)

Fracture zonePseudofault (Robb et al., 2005)Drillhole locationDredge sample location Seismic lines

EP

Cuvier Abyssal Plain (CAP) COTZ

Oceaniccrust

Continent-oceantransition zone (COTZ)

Continentalcrust

Seaward-dippingreflectors (SDRs)

Landwardflows

Moho

Figure 2

? ?

Sediment depositionDeep-water Shallow-water

Deep-marine Sub-aerialLava emplacement

Lava flows Dykes

Oceaniccrust

Continentalcrust

Increasing proportion of igneous materialDecrease in crustal chemical signature

Magmaticcrust

Heavily intrudedcontinental crust

COTZ

DykesInferred boundary between sheeted

dykes and underlying intrusions

Rhaetian

Sinemurian

Pliensbachian

Toarcian

AalenianBajocian

BathonianCallovianOxfordian

Tithonian

Kimmeridgian

Berriasian

Valanginian

Hauterivian

Barremian

Aptian

Albian

Cenomanian

Hettangian

Upp

erLo

wer

Mid

dle

Upp

erLo

wer

Upp

er

Tria

ssic

Jura

ssic

Cre

tace

ous

Exmouth Plateau Cuvier Margin Stratigraphy Tectonic Events

Ma

GearleSiltstone

WindaliaRadiolarite

MuderongShale

BirdrongSandstone

DingoClaystone

AtholFormation

MuratSiltstone

North Rankin Formation

BrigadierFormation

Learmonth Form

ation

Sandstone

Siltstone

Sandstone/Siltstone

Sandstone/Claystone

Radiolarite Terrestrial

Claystone

FluviodeltaicShallow Marine (Clastic)

Offshore Marine (Clastic)

Eroded/condensed

Lithology Depositional environment

Offshore Marine (Low sediment input)

200

180

160

140

120

100

Chronostratigraphicage

BarrowGroup

Winning G

roup

Rifting

Rifting

Gascoyne Cuvier

CO

TZ

CO

TZ

Break-up

Break-up

(A) (B)

200

300

400

500

600

700

T.D. 746

Dep

th (m

)

Claystone

Birdrong Sandstone

Siltstone/Claystone

Siltstone

Shallowest recordedHyperamina spp.

DSDP 263 Pendock-1

800

850

Figure 3

Figure 426

°S24

°S22

°S20

°S18

°S16

°S

108°E 110°E 112°E

0 100 200 300 km-300 0 300Magnetic anomaly (nT)

N

EMAG2v2

108°E 110°E 112°E

0 100 200 300 km-300 0 300Magnetic anomaly (nT)

N

EMAG2v3

108°E 110°E 112°EShiptrack

+ve-ve

Magnetic anomaly polarity

0750 Magnetic anomaly

amplitude (nT)

0 100 200 300 km

N

NCB

SCBPB

WP

WS

CAP

GAP

GP

Cu

Previously defined COTZ

12

10

8

6

4

10 kmTwo-

way

tim

e (s

)

12

10

8

6

4

10 kmTwo-

way

tim

e (s

)

NW SE

?

SonneRidge

(D)

(C)

M5rM6n M10n

M9rM9n

M8rM8n

M6rM7n

M7rM6n

M5rM5n

M10r M10Nr?M10N?

131.7131.7131.9131.4131.4

132.1132.3

132.6132.8

133.1133.3

133.6133.9

Normal polarityReverse polarity

500250

0-250-500M

ag. a

nom

aly

(nT)

Shiptrack magnetic profile [42-41]

10 km

48

12

1620

Dep

th (k

m)

(B)

10 km

10 km

NW SE

468

101214161820

Dep

th (k

m)

468

101214161820

Dep

th (k

m)

0 4 8Interpreted velocity (km s-1)

M5rM6n

M6rM7n

M7rM8n

M8rM9n

M9rM10n

M10r M10Nr?M10N?

131.7131.9

132.1132.3

132.6132.8

133.1133.3

133.6133.9 Magnetic chron ages (Ma)

Normal polarityReverse polarity

200100

0-100-200

Mag

. ano

mal

y (n

T)

10 km

10 km

NW SE

468

101214161820

Dep

th (k

m)

468

101214161820

Dep

th (k

m)

0 4 8Interpreted velocity (km s-1)

(A)

Vertical exaggeration 2x

Fig. 6

Shiptrack magnetic profile [39]

Magnetic chron ages (Ma)

Vertical exaggeration 2x

SedimentarycoverSF1SF2

CO

TZ crust

SF3Upper mantlePredicted extinct spreading ridge (Gibbons et al., 2012)

Continental crustSDR extent -thick end points to dipand divergencedirection

Legend for A, B, D

s135-s135_11 intersection

108°E 110°E 112°EShiptrack

+ve-ve

Magnetic anomaly polarity

0750 Magnetic anomaly

amplitude (nT)

0 100 200 300 km

N

[42]

[39][41]

24°S

22°S

20°S

Fig. 5AFig. 5B

Fig. 5D

Seismic line presentedShiptrack profile used

?

Figure 5

Landward-dipping reflections

Moho

Moho

Moho

Reflections diverge towards Sonne Ridge

10 km

7

8

9

10

11

12

13

14

Dep

th (k

m)

10 km

7

8

9

10

11

12

13

14

Dep

th (k

m)

Sedimentary section

SDR-bearing crust

Opaque crust

Reflective crust

Vertical exaggeration 2x

Figure 6

10 km

3

5

7

9

11

13

Two-

way

Tim

e (s

)

10 km

3

5

7

9

11

13

Two-

way

Tim

e (s

)SW NEs135-s135_05s135-s135_108 repro-n303 EW0113-6 EW0113-5

Wallaby Plateau Sonne Ridge Cape RangeFracture

Zone

Multiples ?

Exmouth Plateau

Figure 7

Moho

10 km10

8

6

4

Two-

way

tim

e (s

)

12

14

10

8

6

4

Two-

way

tim

e (s

)NW SE

12

14

10

8

6

4

Two-

way

tim

e (s

)

10 km

10 km

(A)

? ?

Sonne RidgeWallaby Saddle Previously defined COTZ

NW SE(B)

10 km10

8

6

4

2

Two-

way

tim

e (s

)

10 km10

8

6

4

2

Two-

way

tim

e (s

)

??

Wallaby Saddle

NW SE(C)

10 km10

8

6

4

Two-

way

tim

e (s

)

CDP CDP CDP

Interval velocity (km s-1)

2.5 3.5 4.5 5.5 6.5 7.5

??

0

01 -6

100 200 km

N

Elevation (km)

(D)

108°E 110°E

26°S

24°S

22°S

Fig. 8A

Wallaby Saddle

Wallaby Plateau

Wallaby Plateau

Wallaby Plateau

Multiples

Multiples

Figure 8

Fig. 8C

Fig. 8B

Dipping and diverging reflection packages

Dipping and diverging reflection packages

Moho

Moho

Reducedimaging

-7-5-3-113579

111315

0.702 0.703 0.704 0.705 0.706 0.707

1

10

100

1000Sa

mpl

e/Pr

imiti

ve M

antle

Rb Ba Th Nb U K La Ce Pb Nd Sr Sm Zr Eu Ti Gd Tb Dy Er Yb

Sonne Ridge (057DR051A)Wallaby Plateau A (055BS004A)Wallaby Plateau B (055BS004B)All MORBEnriched-MORBNormal MORBDepleted MORBGLOSSAve. continental crust

87Sr/86Sr

ε(N

d)

057DR051A055BS004A055BS004BAtlantic MORBPacific MORBIndian MORB

(A) (B)

?

N200 km

CRFZ

?

N200 km

CRFZ

N200 km

CRFZ

(B)

(C) (D)

Continental crust COTZ crust Oceanic crust

M10NM10M9M8

M7M6M5nSpreading ridge (active)

Spreading ridge (abandoned)Ridge jump boundaries (pseudofaults)

~134 Ma

~132 Ma ~130 Ma

108°E 110°E 112°E

22°S

20°S

18°S

16°S

200 km-300 0 300

Magnetic anomaly (nT)

Proposed COTZ

(A)

GM

GMGM

CM

CM

CM

SR

SjR

SR SR

Key for B-D

SR

SjR

PreviousCOB

Oldest CAPoceanic crust